數位內容時代的來臨，影像壓縮持續研究的課題著重於如何提升圖片與視訊的壓縮率。在一味地追求壓縮率下，往往失去了圖片或影片的品質。在失真壓縮上，研究不斷地在找尋壓縮率及壓縮品質的平衡。難道無法二者兼顧？在此我們提出的方法，使用在既有的影像壓縮架構之下(在此為H.264 codec)，除了一般性的目的：提高壓縮率，我們在編碼器端，將輸入視訊資料中背景紋理部份加以分析；移除掉背景紋理。並於解碼端我們再將紋理背景合成出來補回，藉此將壓縮率往上提升，且產生在視覺效果上不降低其品質。

　　在本論文中，針對紋理部分佔據一定比例以上以圖片（影片）提出一個有效的演算法。首先，我們對物件做切割，再以顏色，紋理特徵分析我們所到取代掉的背景紋理。剔除太過破碎，面積過小的區域。保留下紋理特徵，顏色位置等參數。於解碼端，除了舊有的參數合成方法可將紋理背景合成補回，我們提出另一合成方式，新的合成方法可更有效果地解決結構性較強的紋理背景。藉此提升取代後的效果。

With age of digital-content coming, the issue from the past has continued to study have focused on how to improve picture and video compression. In the pursuit of compression ratio, the image or video is often lost visual quality. In distortion compression, the researcher pursuit balance between compression ratio and visual quality. Do not balance the two? Here, we present a method use of the existing video compression framework (H.264 codec), in addition to the general target: improved compression ratio. In encoder terminal, the input video background texture are analyzed, and then be removed. In the decoder terminal, texture is synthesized to retrieve the background. By this way, we could increase the compression ration and produce better visual quality.
In this thesis, we present an effective algorithm for texture occupied certain percentage area image or video. First, we segment texture in the image, and then to analysis color, texture characteristic, replace the background texture. Continuing, we remove broken, small region. Reservations texture characteristics, such as location, colors and so on. In the decoder, we propose another synthesis method. In the past, the most use parameters to synthesize the background texture. In addition to the synthesis method can more effectively synthesize the structural background texture. This will increase effectiveness after using texture replaced.

[1] A. Dumitras¸ and B. G. Haskell, “A texture replacement method at the encoder for bit-rate reduction of compressed video,”IEEE Trans. Circuits Syst. Video Technol., vol. 13, pp. 163–175, Feb. 2003.
[2] Dumitras, A.; Haskell, B.G.; “An encoder-decoder texture replacement method with application to content-based movie coding,” IEEE Trans. Circuits Syst. Video Technol., vol. 14, Issue 6, pp. 825–840, Feb. 2004
[3] Ndjiki-Nya, P.; Makai, B.; Blattermann, G.; Smolic, A.; Schwarz, H.; Wiegand, T. “A Content-Base Video Coding Approach for Rigid and Non-Rigid Textures” Digital Object Identifier 8-11 Oct. 2006 Page(s):3169 - 3172 10.1109/ICIP.2006
[4 ] Ndjiki-Nya, P.; Hinz, T.; Smolic, A.; Wiegand, T.; ”A generic and automatic content-based approach for improved H.264/MPEG4-AVC video coding” Volume 2, 11-14 Sept. 2005 Page(s):II - 874-7 Digital Object Identifier 10.1109/ICIP.2005.1530195
[5] Ndjiki-Nya, P.; Kootz, M.; Wiegand, T.; “Automatic detection of video synthesis related artifacts” Digital Object Identifier 17-21 May 2004 Page(s):iii - 733-6 vol.3
10.1109/ICASSP.2004.1326649
[6] Ndjiki-Nya, P.; Makai, B.; Blattermann, G.; Smolic, A.; Schwarz, H.; Wiegand, T. “Improved H.264/AVC coding using texture analysis and synthesis.” Digital Object Identifier 10.1109/ICIP.2003.1247378, Volume 3, 14-17, Page(s):III - 849-52, Sept. 2003
[7] J. and E. P. Simoncelli, “A parametric texture model based on joint statistics of complex wavelet coefficients,” Int. J. Comput. Vis., vol. 40, no. 1, pp. 49–71, 2000.
[8] Deng, Y.; Manjunath, B.S.; “Unsupervised segmentation of color-texture regions in images and video” IEEE Trans. Digital Object Identifier., vol. 23, Issue 8, pp. 800–810, Feb. 2001
[9] Manjunath, B.S., Wu, P., Newsam, S., and Shin, H.D. “A Texture Descriptor for Browsing and Similarity Retrieval.” Journal of Signal Processing: Image Communication, Volume 16, Issue 1-2, page 33-43, Sept 2000.